Increasing venoarterial extracorporeal membrane oxygenation flow puts higher demands on left ventricular work in a porcine model of chronic heart failure.

BACKGROUND: Venoarterial extracorporeal membrane oxygenation (VA ECMO) is widely used in the treatment of circulatory failure, but repeatedly, its negative effects on the left ventricle (LV) have been observed. The purpose of this study is to assess the influence of increasing extracorporeal blood flow (EBF) on LV performance during VA ECMO therapy of decompensated chronic heart failure. METHODS: A porcine model of low-output chronic heart failure was developed by long-term fast cardiac pacing. Subsequently, under total anesthesia and artificial ventilation, VA ECMO was introduced to a total of five swine with profound signs of chronic cardiac decompensation. LV performance and organ specific parameters were recorded at different levels of EBF using a pulmonary artery catheter, a pressure-volume loop catheter positioned in the LV, and arterial flow probes on systemic arteries. RESULTS: Tachycardia-induced cardiomyopathy led to decompensated chronic heart failure with mean cardiac output of 2.9 ± 0.4 L/min, severe LV dilation, and systemic hypoperfusion. By increasing the EBF from minimal flow to 5 L/min, we observed a gradual increase of LV peak pressure from 49 ± 15 to 73 ± 11 mmHg (P = 0.001) and an improvement in organ perfusion. On the other hand, cardiac performance parameters revealed higher demands put on LV function: LV end-diastolic pressure increased from 7 ± 2 to 15 ± 3 mmHg, end-diastolic volume increased from 189 ± 26 to 218 ± 30 mL, end-systolic volume increased from 139 ± 17 to 167 ± 15 mL (all P < 0.001), and stroke work increased from 1434 ± 941 to 1892 ± 1036 mmHg*mL (P < 0.05). LV ejection fraction and isovolumetric contractility index did not change significantly. CONCLUSIONS: In decompensated chronic heart failure, excessive VA ECMO flow increases demands and has negative effects on the workload of LV. To protect the myocardium from harm, VA ECMO flow should be adjusted with respect to not only systemic perfusion, but also to LV parameters.

Microcirculatory blood flow during cardiac arrest and cardiopulmonary resuscitation does not correlate with global hemodynamics: an experimental study.

BACKGROUND: Current research highlights the role of microcirculatory disorders in post-cardiac arrest patients. Affected microcirculation shows not only dissociation from systemic hemodynamics but also strong connection to outcome of these patients. However, only few studies evaluated microcirculation directly during cardiac arrest (CA) and cardiopulmonary resuscitation (CPR). The aim of our experimental study in a porcine model was to describe sublingual microcirculatory changes during CA and CPR using recent videomicroscopic technology and provide a comparison to parameters of global hemodynamics. METHODS: Cardiac arrest was induced in 18 female pigs (50 ± 3 kg). After 3 min without treatment, 5 min of mechanical CPR followed. Continuous hemodynamic monitoring including systemic blood pressure and carotid blood flow was performed and blood lactate was measured at the end of baseline and CPR. Sublingual microcirculation was assessed by the Sidestream Dark Field (SDF) technology during baseline, CA and CPR. Following microcirculatory parameters were assessed off-line separately for capillaries (≤20 µm) and other vessels: total and perfused vessel density (TVD, PVD), proportion of perfused vessels (PPV), microvascular flow index (MFI) and heterogeneity index (HI). RESULTS: In comparison to baseline the CA small vessel microcirculation was only partially preserved: TVD 15.64 (13.59-18.48) significantly decreased to 12.51 (10.57-13.98) mm/mm(2), PVD 15.57 (13.56-17.80) to 5.53 (4.17-6.60) mm/mm(2), PPV 99.64 (98.05-100.00) to 38.97 (27.60-46.29) %, MFI 3.00 (3.00-3.08) to 1.29 (1.08-1.58) and HI increased from 0.08 (0.00-0.23) to 1.5 (0.71-2.00), p = 0.0003 for TVD and <0.0001 for others, respectively. Microcirculation during ongoing CPR in small vessels reached 59-85 % of the baseline values: TVD 13.33 (12.11-15.11) mm/mm(2), PVD 9.34 (7.34-11.52) mm/mm(2), PPV 72.34 (54.31-87.87) %, MFI 2.04 (1.58-2.42), HI 0.65 (0.41-1.07). The correlation between microcirculation and global hemodynamic parameters as well as to lactate was only weak to moderate (i.e. Spearman's ρ 0.02-0.51) and after adjustment for multiple correlations it was non-significant. CONCLUSIONS: Sublingual microcirculatory parameters did not correlate with global hemodynamic parameters during simulated porcine model of CA and CPR. SDF imaging provides additional information about tissue perfusion in the course of CPR.

Inducibility of ventricular fibrillation during mild therapeutic hypothermia: electrophysiological study in a swine model.

INTRODUCTION: Mild therapeutic hypothermia (MTH) is being used after cardiac arrest for its expected improvement in neurological outcome. Safety of MTH concerning inducibility of malignant arrhythmias has not been satisfactorily demonstrated. This study compares inducibility of ventricular fibrillation (VF) before and after induction of MTH in a whole body swine model and evaluates possible interaction with changing potassium plasma levels. METHODS: The extracorporeal cooling was introduced in fully anesthetized swine (n = 6) to provide MTH. Inducibility of VF was studied by programmed ventricular stimulation three times in each animal under the following: during normothermia (NT), after reaching the core temperature of 32°C (HT) and after another 60 minutes of stable hypothermia (HT60). Inducibility of VF, effective refractory period of the ventricles (ERP), QTc interval and potassium plasma levels were measured. RESULTS: Starting at normothermia of 38.7 (IQR 38.2; 39.8)°C, HT was achieved within 54 (39; 59) minutes and the core temperature was further maintained constant. Overall, the inducibility of VF was 100% (18/18 attempts) at NT, 83% (15/18) after reaching HT (P = 0.23) and 39% (7/18) at HT60 (P = 0.0001) using the same protocol. Similarly, ERP prolonged from 140 (130; 150) ms at NT to 206 (190; 220) ms when reaching HT (P < 0.001) and remained 206 (193; 220) ms at HT60. QTc interval was inversely proportional to the core temperature and extended from 376 (362; 395) at NT to 570 (545; 599) ms at HT. Potassium plasma level changed spontaneously: decreased during cooling from 4.1 (3.9; 4.8) to 3.7 (3.4; 4.1) mmol/L at HT (P < 0.01), then began to increase and returned to baseline level at HT60 (4.6 (4.4; 5.0) mmol/L, P = NS). CONCLUSIONS: According to our swine model, MTH does not increase the risk of VF induction by ventricular pacing in healthy hearts. Moreover, when combined with normokalemia, MTH exerts an antiarrhythmic effect despite prolonged QTc interval.

Increasing venoarterial extracorporeal membrane oxygenation flow negatively affects left ventricular performance in a porcine model of cardiogenic shock.

BACKGROUND: The aim of this study was to assess the relationship between extracorporeal blood flow (EBF) and left ventricular (LV) performance during venoarterial extracorporeal membrane oxygenation (VA ECMO) therapy. METHODS: Five swine (body weight 45 kg) underwent VA ECMO implantation under general anesthesia and artificial ventilation. Subsequently, acute cardiogenic shock with signs of tissue hypoxia was induced. Hemodynamic and cardiac performance parameters were then measured at different levels of EBF (ranging from 1 to 5 L/min) using arterial and venous catheters, a pulmonary artery catheter and a pressure-volume loop catheter introduced into the left ventricle. RESULTS: Myocardial hypoxia resulted in a decline in mean (±SEM) cardiac output to 2.8 ± 0.3 L/min and systolic blood pressure (SBP) to 60 ± 7 mmHg. With an increase in EBF from 1 to 5 L/min, SBP increased to 97 ± 8 mmHg (P < 0.001); however, increasing EBF from 1 to 5 L/min significantly negatively influences several cardiac performance parameters: cardiac output decreased form 2.8 ± 0.3 L/min to 1.86 ± 0.53 L/min (P < 0.001), LV end-systolic volume increased from 64 ± 11 mL to 83 ± 14 mL (P < 0.001), LV stroke volume decreased from 48 ± 9 mL to 40 ± 8 mL (P = 0.045), LV ejection fraction decreased from 43 ± 3 % to 32 ± 3 % (P < 0.001) and stroke work increased from 2096 ± 342 mmHg mL to 3031 ± 404 mmHg mL (P < 0.001). LV end-diastolic pressure and volume were not significantly affected. CONCLUSIONS: The results of the present study indicate that higher levels of VA ECMO blood flow in cardiogenic shock may negatively affect LV function. Therefore, it appears that to mitigate negative effects on LV function, optimal VA ECMO blood flow should be set as low as possible to allow adequate tissue perfusion.

Tachycardia-Induced Cardiomyopathy As a Chronic Heart Failure Model in Swine.

A stable and reliable model of chronic heart failure is required for many experiments to understand hemodynamics or to test effects of new treatment methods. Here, we present such a model by tachycardia-induced cardiomyopathy, which can be produced by rapid cardiac pacing in swine. A single pacing lead is introduced transvenously into fully anaesthetized healthy swine, to the apex of the right ventricle, and fixated. Its other end is then tunneled dorsally to the paravertebral region. There, it is connected to an in-house modified heart pacemaker unit that is then implanted in a subcutaneous pocket. After 4 - 8 weeks of rapid ventricular pacing at rates of 200 - 240 beats/min, physical examination revealed signs of severe heart failure - tachypnea, spontaneous sinus tachycardia, and fatigue. Echocardiography and X-ray showed dilation of all heart chambers, effusions, and severe systolic dysfunction. These findings correspond well to decompensated dilated cardiomyopathy and are also preserved after the cessation of pacing. This model of tachycardia-induced cardiomyopathy can be used for studying the pathophysiology of progressive chronic heart failure, especially hemodynamic changes caused by new treatment modalities like mechanical circulatory supports. This methodology is easy to perform and the results are robust and reproducible.